Numerical analysis of cartilage surfaces for osteoarthritis diagnosis using field and feature parameters
Introduction
Osteoarthritis (OA) is a worldwide articular cartilage (AC) progressive wear disease which affects majority of people under certain age group [1]. OA disease is also one of the top ranked disabling diseases and the treatment cost plus work-related losses is over $33 billion annually in United States [2]. AC surface fibrillation and collagen matrix losing integrity sometimes with bone exposure are major symptoms of OA disease [3], [4]. Based on the erosion of total AC loss, OA could be classified into three degrees from OA grade 1–3 which stands for the severity of different disease stages [5].
AC is a connective tissue and bony component located on the end of a synovial joint such as a knee joint. It is composed of different layer structures which are superficial layer, intermediate collagen network and calcified subchondral bone from surface to bottom [6], [7]. Existing studies have revealed that the cartilage surface texture changes with OA progression [8], [9], [10], and therefore can be used to assess OA conditions.
To study the cartilage surface texture change, laser scanning confocal microscope (LSCM) could be employed as a powerful tool for target topography assessment. Compared with traditional scanning electronic microscope, LSCM has significant advantages in terms of zero or very limited sample preparation requirements and its capability of acquiring three-dimensional information. Other than two-dimensional information from a lateral x–y dimensions planar AC surface, LSCM could gather serial optical sections through z-direction step in scanning from a controlled field depth [11], [12], [13]. Appropriate 3D images collected through LSCM could be numerically analyzed with surface texture parameters to study the AC surface changes with OA progression.
Field and feature parameters are two defined sets of surface texture parameters compiled by ISO/FDIS 25178-2 [14] (the international standard about areal characterization of surface textures). The field parameters apply statistics to the areal data (cloud of points) of a surface texture, which include the majority of the conventional 3D surface texture parameters. The feature parameters apply statistics from a subset of pre-defined topographic features [15]. This characterisation method adopts feature recognition techniques, and it can prune out those unstable features [16], and it is expected to be more stable than the conventional surface texture parameters.
All of the existing approaches use statistical descriptions (i.e., the traditional shape and profile texture parameters) to monitor changes in the particle shape and surface texture to indicate changes in the wear conditions of the joint. The parameters cannot assess functional performance of the joint directly, and therefore cannot be used as a diagnostic tool to study the cause of wear problems [16]. Wear in joints depends not only on the surface morphology of the counter-part cartilages which can be characterised using the above areal (field) parameters in 3D, but is also determined by their contact areas. To date, researchers have difficulty in measuring or predicting the contact areas between two wear components. Feature parameters have been developed to “characterise the functional topographical features of surfaces by analyzing texture shape and direction, estimation of feature attributes and differentiation between connected and isolated features” [17]. This new concept has opened a door for the study of OA causes and functional performance of the cartilage. Unique to the existing research, this study investigates the newly developed functional parameters for OA assessment.
Section snippets
Wear testing and image acquisition
3D LSCM imaging AC surface method has been employed and recognized as an effective way for OA assessment [18], [19]. Sheep joints were wear tested using a wear simulator to obtain cartilage samples with the defined degrees of OA. Specimen pieces were collected from tibia end, then fixed and stained for LSCM imaging. Images containing 3D surface information were numerically analyzed using field and feature parameters. Statistical and correlation analysis was performed to examine the new
Filtering
Practical biomedical surfaces are usually multi-scale in nature, i.e. they are comprised of various topographic features with distinct scales. The cartilages also have the composite surface texture with multi-scale features. For instance, as shown in Fig. 2, a representative surface texture of Joint 3 clearly presents two dominant feature components with differing wavelengths. One is the large scale feature which has the wavelength larger than 50 μm while the other component is the micro-scale
Discussion
In this study, the field and feature parameters defined in ISO/FDIS 25178-2 have been employed to study the surface texture of cartilage samples collected from sheep knee joints. Six groups of AC samples were collected from health knee joint to OA degrees 1, 2 and 3. It is the first time that the feature parameters have been applied together with the field parameters to the AC wear studies.
After the one way ANOVA analysis, 24 parameters out of total 32 have returned a high F value showing
Conclusion
This study has used the field and feature parameters from ISO/FDIS 25178 for analyzing cartilage surface texture changes under the degrees of wear conditions varying from normal condition to OA grade 3. Seven field parameters and three feature parameters have been selected as critical numerical descriptors to characterise the surface evolution in the study. The trend of the field parameters is similar to what has been reported in the previous studies.
This is the first time the feature
Acknowledgements
The authors would like to acknowledge Australian Research Council (ARC) research grant DP1093975 for its support to this project.
References (24)
- et al.
MR imaging findings in early osteoarthritis of the knee
European Journal of Radiology
(2004) - et al.
Quantitative assessment of cartilage surface roughness in osteoarthritis using high frequency ultrasound
Ultrasound in Medicine & Biology
(1992) - et al.
145 Scratching the surface: structural and functional changes of the articular cartilage surface can be measured with atomic force microscopy in an experimental model of early osteoarthritis
- et al.
The effects of lipid depletion on osteoarthritic wear
Wear
(2002) - et al.
The development of three-dimensional imaging techniques of wear particle analysis
Wear
(1997) Feature parameters
Wear
(2009)Osteoarthritis diagnosis using wear particle analysis technique: investigation of correlation between particle and cartilage surface in walking process
Wear
(2007)The influence of pathology, pain, balance, and self-efficacy on function in women with osteoarthritis of the knee
Physical Therapy
(2004)- et al.
Robbins and Cotran Pathologic Basis of Disease
(2009) Adult Articular Cartilage
(1973)
Articular Cartilage and Knee Joing Function, Basic Science and Arthroscopy
Joint Structure & Function, A Comprehensive Analysis
Cited by (12)
3D characterisation of surface topographies of wear debris collected from human knees for osteoarthritis study
2019, Tribology InternationalCitation Excerpt :To differentiate the normal and severe OA conditions, three functional parameters demonstrated their effectiveness. Sdc0_5 can give an indication of the optimum crevice volume which is not too deep to prevent leakage for lubricant entrapment [37]. Since Sdc0_5 reveals the height differences between the maximum point to 5% of the bearing area ratios, Sdc0_5 indicates the entrapment ability of the wear debris surface.
Quantification of cartilage wear morphologies in unidirectional sliding experiments: Influence of different macromolecular lubricants
2017, BiotribologyCitation Excerpt :As discussed above, a simple parameter for charaterizing the topography of a surface is the roughness, which can be quantified either by the root mean square roughness (Sq) or the average roughness (Sa). Literature values on the surface roughness of articular cartilage vary a lot between species and between different studies: Sa = 0.07 μm (human) [61], Sa = 0.18–0.26 μm (bovine) [62], Sq = 2.34 μm (sheep) [63], Sq = 5.32 μm (human) [2]. Furthermore, there are a few studies which quantified cartilage based on a two dimensional roughness value, i.e. Ra, which is basically the 2D version of the Sa value.
Surface texture metrology for metal additive manufacturing: a review
2016, Precision EngineeringCitation Excerpt :The majority of existing reference examples where areal characterisation has been used employ ISO 25178-2 texture parameters which are the direct counterpart of the ISO 4287 profile parameters. ISO 25178-2 feature parameters, which could help a great deal at isolating surface areas of interest [120] have not been explored in the literature on surface metrology for AM. Measurement and characterisation is often not fully reproducible as key information is not reported (for example, void treatment, reduction of measurement anomalies, levelling, filtering and sample spacing).
Wear characterisation of articular cartilage surfaces at a nano-scale using atomic force microscopy
2013, Tribology InternationalCitation Excerpt :They are capable of characterising functional properties, for example examining the cutting performance of grinding wheels [5]. The field and feature parameters have been successfully applied to characterise the surface topography of worn surfaces such as matt finish femoral stems [5], TiN coated plates in total knee implants [5], hard on hard bearings in prosthetic hip joints [5] and cartilage surface at a micrometre scale [3]. The above approaches have not, however, been applied to investigate worn cartilage surfaces at a nanometre scale.